Search for a light Higgs resonance in radiative decays
of the ϒð1SÞ with a charm tag
J. P. Lees,1 V. Poireau,1 V. Tisserand,1 E. Grauges,2 A. Palano,3a,3b G. Eigen,4 B. Stugu,4 D. N. Brown,5 L. T. Kerth,5 Yu. G. Kolomensky,5M. J. Lee,5 G. Lynch,5 H. Koch,6T. Schroeder,6 C. Hearty,7T. S. Mattison,7 J. A. McKenna,7 R. Y. So,7A. Khan,8V. E. Blinov,9a,9b,9cA. R. Buzykaev,9aV. P. Druzhinin,9a,9bV. B. Golubev,9a,9bE. A. Kravchenko,9a,9b A. P. Onuchin,9a,9b,9c S. I. Serednyakov,9a,9bYu. I. Skovpen,9a,9b E. P. Solodov,9a,9b K. Yu. Todyshev,9a,9b A. J. Lankford,10 B. Dey,11J. W. Gary,11O. Long,11M. Franco Sevilla,12T. M. Hong,12D. Kovalskyi,12J. D. Richman,12C. A. West,12
A. M. Eisner,13W. S. Lockman,13W. Panduro Vazquez,13B. A. Schumm,13A. Seiden,13 D. S. Chao,14C. H. Cheng,14 B. Echenard,14K. T. Flood,14D. G. Hitlin,14T. S. Miyashita,14P. Ongmongkolkul,14F. C. Porter,14M. Röhrken,14 R. Andreassen,15Z. Huard,15B. T. Meadows,15B. G. Pushpawela,15M. D. Sokoloff,15L. Sun,15P. C. Bloom,16W. T. Ford,16 A. Gaz,16J. G. Smith,16S. R. Wagner,16R. Ayad,17,†W. H. Toki,17B. Spaan,18D. Bernard,19M. Verderi,19S. Playfer,20 D. Bettoni,21aC. Bozzi,21a R. Calabrese,21a,21bG. Cibinetto,21a,21bE. Fioravanti,21a,21b I. Garzia,21a,21b E. Luppi,21a,21b
L. Piemontese,21a V. Santoro,21a A. Calcaterra,22R. de Sangro,22G. Finocchiaro,22S. Martellotti,22P. Patteri,22 I. M. Peruzzi,22,‡ M. Piccolo,22 M. Rama,22A. Zallo,22R. Contri,23a,23bM. R. Monge,23a,23b S. Passaggio,23a C. Patrignani,23a,23b B. Bhuyan,24 V. Prasad,24A. Adametz,25U. Uwer,25H. M. Lacker,26U. Mallik,27C. Chen,28
J. Cochran,28S. Prell,28H. Ahmed,29A. V. Gritsan,30N. Arnaud,31M. Davier,31D. Derkach,31G. Grosdidier,31 F. Le Diberder,31A. M. Lutz,31B. Malaescu,31,§P. Roudeau,31A. Stocchi,31G. Wormser,31D. J. Lange,32D. M. Wright,32
J. P. Coleman,33J. R. Fry,33E. Gabathuler,33D. E. Hutchcroft,33 D. J. Payne,33C. Touramanis,33A. J. Bevan,34 F. Di Lodovico,34R. Sacco,34 G. Cowan,35D. N. Brown,36C. L. Davis,36 A. G. Denig,37M. Fritsch,37W. Gradl,37 K. Griessinger,37A. Hafner,37K. R. Schubert,37R. J. Barlow,38,¶G. D. Lafferty,38R. Cenci,39B. Hamilton,39A. Jawahery,39 D. A. Roberts,39R. Cowan,40R. Cheaib,41P. M. Patel,41,*S. H. Robertson,41N. Neri,42aF. Palombo,42a,42bL. Cremaldi,43
R. Godang,43,** D. J. Summers,43M. Simard,44 P. Taras,44G. De Nardo,45a,45bG. Onorato,45a,45bC. Sciacca,45a,45b G. Raven,46C. P. Jessop,47J. M. LoSecco,47K. Honscheid,48R. Kass,48M. Margoni,49a,49bM. Morandin,49aM. Posocco,49a
M. Rotondo,49a G. Simi,49a,49b F. Simonetto,49a,49bR. Stroili,49a,49bS. Akar,50E. Ben-Haim,50M. Bomben,50G. R. Bonneaud,50H. Briand,50G. Calderini,50J. Chauveau,50Ph. Leruste,50G. Marchiori,50J. Ocariz,50M. Biasini,51a,51b E. Manoni,51aA. Rossi,51aC. Angelini,52a,52bG. Batignani,52a,52bS. Bettarini,52a,52bM. Carpinelli,52a,52b,††G. Casarosa,52a,52b M. Chrzaszcz,52a F. Forti,52a,52bM. A. Giorgi,52a,52b A. Lusiani,52a,52c B. Oberhof,52a,52bE. Paoloni,52a,52bG. Rizzo,52a,52b J. J. Walsh,52aD. Lopes Pegna,53J. Olsen,53A. J. S. Smith,53F. Anulli,54aR. Faccini,54a,54bF. Ferrarotto,54aF. Ferroni,54a,54b M. Gaspero,54a,54b A. Pilloni,54a,54bG. Piredda,54a C. Bünger,55S. Dittrich,55O. Grünberg,55 M. Hess,55T. Leddig,55
C. Voß,55R. Waldi,55 T. Adye,56E. O. Olaiya,56F. F. Wilson,56S. Emery,57G. Vasseur,57D. Aston,58D. J. Bard,58 C. Cartaro,58 M. R. Convery,58J. Dorfan,58G. P. Dubois-Felsmann,58W. Dunwoodie,58M. Ebert,58R. C. Field,58 B. G. Fulsom,58M. T. Graham,58C. Hast,58W. R. Innes,58P. Kim,58 D. W. G. S. Leith,58D. Lindemann,58 S. Luitz,58
V. Luth,58H. L. Lynch,58D. B. MacFarlane,58D. R. Muller,58 H. Neal,58M. Perl,58,* T. Pulliam,58B. N. Ratcliff,58 A. Roodman,58R. H. Schindler,58A. Snyder,58D. Su,58M. K. Sullivan,58J. Va’vra,58W. J. Wisniewski,58H. W. Wulsin,58
M. V. Purohit,59J. R. Wilson,59A. Randle-Conde,60S. J. Sekula,60M. Bellis,61P. R. Burchat,61E. M. T. Puccio,61 M. S. Alam,62J. A. Ernst,62R. Gorodeisky,63N. Guttman,63D. R. Peimer,63A. Soffer,63S. M. Spanier,64J. L. Ritchie,65
R. F. Schwitters,65J. M. Izen,66X. C. Lou,66F. Bianchi,67a,67b F. De Mori,67a,67bA. Filippi,67a D. Gamba,67a,67b L. Lanceri,68a,68bL. Vitale,68a,68bF. Martinez-Vidal,69A. Oyanguren,69P. Villanueva-Perez,69J. Albert,70Sw. Banerjee,70
A. Beaulieu,70F. U. Bernlochner,70 H. H. F. Choi,70G. J. King,70R. Kowalewski,70M. J. Lewczuk,70T. Lueck,70 I. M. Nugent,70J. M. Roney,70R. J. Sobie,70N. Tasneem,70T. J. Gershon,71P. F. Harrison,71T. E. Latham,71H. R. Band,72
S. Dasu,72Y. Pan,72R. Prepost,72and S. L. Wu72 (BABAR Collaboration)
1Laboratoire d’Annecy-le-Vieux de Physique des Particules (LAPP), Université de Savoie, CNRS/IN2P3, F-74941 Annecy-Le-Vieux, France
2Universitat de Barcelona, Facultat de Fisica, Departament ECM, E-08028 Barcelona, Spain 3aINFN Sezione di Bari, I-70126 Bari, Italy
3bDipartimento di Fisica, Università di Bari, I-70126 Bari, Italy 4University of Bergen, Institute of Physics, N-5007 Bergen, Norway
5Lawrence Berkeley National Laboratory and University of California, Berkeley, California 94720, USA 6Ruhr Universität Bochum, Institut für Experimentalphysik 1, D-44780 Bochum, Germany
7University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z1 8Brunel University, Uxbridge, Middlesex UB8 3PH, United Kingdom 9aBudker Institute of Nuclear Physics SB RAS, Novosibirsk 630090, Russia
9cNovosibirsk State Technical University, Novosibirsk 630092, Russia 10University of California at Irvine, Irvine, California 92697, USA 11University of California at Riverside, Riverside, California 92521, USA 12University of California at Santa Barbara, Santa Barbara, California 93106, USA
13University of California at Santa Cruz, Institute for Particle Physics, Santa Cruz, California 95064, USA 14California Institute of Technology, Pasadena, California 91125, USA
15University of Cincinnati, Cincinnati, Ohio 45221, USA 16University of Colorado, Boulder, Colorado 80309, USA 17Colorado State University, Fort Collins, Colorado 80523, USA
18Technische Universität Dortmund, Fakultät Physik, D-44221 Dortmund, Germany 19Laboratoire Leprince-Ringuet, Ecole Polytechnique, CNRS/IN2P3, F-91128 Palaiseau, France
20University of Edinburgh, Edinburgh EH9 3JZ, United Kingdom 21aINFN Sezione di Ferrara, I-44122 Ferrara, Italy
21bDipartimento di Fisica e Scienze della Terra, Università di Ferrara, I-44122 Ferrara, Italy 22INFN Laboratori Nazionali di Frascati, I-00044 Frascati, Italy
23aINFN Sezione di Genova, I-16146 Genova, Italy
23bDipartimento di Fisica, Università di Genova, I-16146 Genova, Italy 24Indian Institute of Technology Guwahati, Guwahati, Assam 781 039, India 25Universität Heidelberg, Physikalisches Institut, D-69120 Heidelberg, Germany 26Humboldt-Universität zu Berlin, Institut für Physik, D-12489 Berlin, Germany
27University of Iowa, Iowa City, Iowa 52242, USA 28Iowa State University, Ames, Iowa 50011-3160, USA
29Physics Department, Jazan University, Jazan 22822, Kingdom of Saudi Arabia 30Johns Hopkins University, Baltimore, Maryland 21218, USA
31Laboratoire de l’Accélérateur Linéaire, IN2P3/CNRS et Université Paris-Sud 11, Centre Scientifique d’Orsay, F-91898 Orsay Cedex, France
32Lawrence Livermore National Laboratory, Livermore, California 94550, USA 33University of Liverpool, Liverpool L69 7ZE, United Kingdom 34Queen Mary, University of London, London E1 4NS, United Kingdom 35University of London, Royal Holloway and Bedford New College, Egham,
Surrey TW20 0EX, United Kingdom
36University of Louisville, Louisville, Kentucky 40292, USA
37Johannes Gutenberg-Universität Mainz, Institut für Kernphysik, D-55099 Mainz, Germany 38University of Manchester, Manchester M13 9PL, United Kingdom
39University of Maryland, College Park, Maryland 20742, USA
40Massachusetts Institute of Technology, Laboratory for Nuclear Science, Cambridge, Massachusetts 02139, USA
41McGill University, Montréal, Québec, Canada H3A 2T8 42aINFN Sezione di Milano, I-20133 Milano, Italy
42bDipartimento di Fisica, Università di Milano, I-20133 Milano, Italy 43University of Mississippi, University, Mississippi 38677, USA
44Université de Montréal, Physique des Particules, Montréal, Québec, Canada H3C 3J7 45aINFN Sezione di Napoli, I-80126 Napoli, Italy
45bDipartimento di Scienze Fisiche, Università di Napoli Federico II, I-80126 Napoli, Italy 46NIKHEF, National Institute for Nuclear Physics and High Energy Physics,
NL-1009 DB Amsterdam, The Netherlands
47University of Notre Dame, Notre Dame, Indiana 46556, USA 48Ohio State University, Columbus, Ohio 43210, USA
49aINFN Sezione di Padova, I-35131 Padova, Italy
49bDipartimento di Fisica, Università di Padova, I-35131 Padova, Italy 50Laboratoire de Physique Nucléaire et de Hautes Energies, IN2P3/CNRS, Université Pierre et Marie Curie-Paris6, Université Denis Diderot-Paris7,
F-75252 Paris, France
51aINFN Sezione di Perugia, I-06123 Perugia, Italy
51bDipartimento di Fisica, Università di Perugia, I-06123 Perugia, Italy 52aINFN Sezione di Pisa, I-56127 Pisa, Italy
52bDipartimento di Fisica, Università di Pisa, I-56127 Pisa, Italy 52cScuola Normale Superiore di Pisa, I-56127 Pisa, Italy 53Princeton University, Princeton, New Jersey 08544, USA
54bDipartimento di Fisica, Università di Roma La Sapienza, I-00185 Roma, Italy 55Universität Rostock, D-18051 Rostock, Germany
56Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX, United Kingdom 57CEA, Irfu, SPP, Centre de Saclay, F-91191 Gif-sur-Yvette, France
58SLAC National Accelerator Laboratory, Stanford, California 94309 USA 59University of South Carolina, Columbia, South Carolina 29208, USA
60Southern Methodist University, Dallas, Texas 75275, USA 61Stanford University, Stanford, California 94305-4060, USA 62State University of New York, Albany, New York 12222, USA 63Tel Aviv University, School of Physics and Astronomy, Tel Aviv, 69978, Israel
64University of Tennessee, Knoxville, Tennessee 37996, USA 65University of Texas at Austin, Austin, Texas 78712, USA 66University of Texas at Dallas, Richardson, Texas 75083, USA
67aINFN Sezione di Torino, I-10125 Torino, Italy
67bDipartimento di Fisica, Università di Torino, I-10125 Torino, Italy 68aINFN Sezione di Trieste, I-34127 Trieste, Italy
68bDipartimento di Fisica, Università di Trieste, I-34127 Trieste, Italy 69IFIC, Universitat de Valencia-CSIC, E-46071 Valencia, Spain 70University of Victoria, Victoria, British Columbia, Canada V8W 3P6 71Department of Physics, University of Warwick, Coventry CV4 7AL, United Kingdom
72University of Wisconsin, Madison, Wisconsin 53706, USA (Received 23 February 2015; published 10 April 2015)
A search is presented for the decay ϒð1SÞ → γA0, A0→ c¯c, where A0 is a candidate for the CP-odd Higgs boson of the next-to-minimal supersymmetric standard model. The search is based on data collected with the BABAR detector at the ϒð2SÞ resonance. A sample of ϒð1SÞ mesons is selected via the decay ϒð2SÞ → πþπ−ϒð1SÞ. The A0→ c¯c decay is identified through the reconstruction of hadronic D0, Dþ, and D$ð2010Þþ meson decays. No significant signal is observed. The measured 90% confidence-level upper limits on the product branching fraction Bðϒð1SÞ → γA0
Þ × BðA0→ c¯cÞ range from 7.4 × 10−5to 2.4 × 10−3for A0masses from 4.00 to 8.95 GeV=c2and 9.10 to 9.25 GeV=c2, where the region between 8.95 and 9.10 GeV=c2is excluded because of background from ϒð2SÞ → γχ
bJð1PÞ, χbJð1PÞ → γϒð1SÞ decays.
DOI:10.1103/PhysRevD.91.071102 PACS numbers: 12.15.Ji, 12.60.Fr, 13.20.Gd, 14.80.Da
The next-to-minimal supersymmetric standard model (NMSSM) is an appealing extension of the standard model (SM). It solves the μ problem of the minimal supersym-metric standard model and the hierarchy problem of the SM
[1,2]. The NMSSM has a rich Higgs sector of two charged,
three neutral CP-even, and two neutral CP-odd bosons. Although the Higgs boson discovered at the CERN LHC
[3,4]is consistent with the SM Higgs boson, it can also be
interpreted as one of the heavier Higgs bosons of the NMSSM [5]. The least heavy of the NMSSM Higgs bosons, denoted A0, could be light enough to be produced
in the decay of an ϒ meson[1,6].
In the context of type I or type II two-Higgs-doublet models, the branching fractions of the A0depend on the A0
mass and the NMSSM parameter tan β [7]. Below the charm mass threshold, the A0preferentially decays into two
gluons if tan β is of order 1, and to s¯s or to μþμ−if tan β is
of order 10. Above the charm mass threshold, the A0decays
mainly to c¯c for tan β of order 1 and to τþτ− for tan β of
order 10. BABAR has already ruled out much of the NMSSM parameter space for A0masses below the charm
mass threshold through searches for A0→ μþμ− [8,9]and
for A0→ gg or s¯s[10]. Above the charm mass threshold,
BABAR has ruled out some of the parameter space for high tan β with the A0→ τþτ− searches [11,12]. None of the
searches from BABAR have observed a significant signal, nor have the searches in leptonic channels from the CMS and CLEO[13–15]Collaborations. The A0→ c¯c channel
is one of the last channels that has not yet been explored. We report a search for the decay ϒð1SÞ → γA0, A0→ c¯c
with A0 masses ranging between 4.00 and 9.25 GeV=c2.
An ϒð1SÞ decay is tagged by the presence of a pion pair from ϒð2SÞ → πþπ−ϒð1SÞ. An A0→ c¯c decay is tagged
by the presence of at least one charmed meson such as a D0, a Dþ, or a D$
ð2010Þþ. Therefore, candidates are
*Deceased.
†Now at University of Tabuk, Tabuk 71491, Saudi Arabia. ‡Also at Università di Perugia, Dipartimento di Fisica, I-06123
Perugia, Italy.
§Now at Laboratoire de Physique Nucléaire et de Hautes
Energies, IN2P3/CNRS, F-75252 Paris, France.
∥Now at University of Huddersfield, Huddersfield HD1 3DH,
United Kingdom.
¶Now at University of South Alabama, Mobile, Alabama
36688, USA.
constructed from the combination of a photon, a D meson, and a dipion candidate. An exclusive reconstruction of the A0 is not attempted. Instead, a search is performed in the
spectrum of the invariant mass of the system that recoils against the dipion-photon system. The analysis is there-fore sensitive to the production of any charm resonance produced in the radiative decays of the ϒð1SÞ meson.
The data were recorded with the BABAR detector at the PEP-II asymmetric-energy eþe− collider at the SLAC
National Accelerator Laboratory. The BABAR detector is described in detail elsewhere[16,17]. We use 13.6 fb−1 of
“on-resonance” data collected at the ϒð2SÞ resonance, corresponding to ð98.3 % 0.9Þ × 106 ϒð2SÞ mesons [18],
which includes an estimated ð17.5 % 0.3Þ × 106ϒð2SÞ →
πþπ−ϒð1SÞ decays[19]. The non-ϒð2SÞ backgrounds are
studied using 1.4 fb−1 of “off-resonance” data collected
30 MeV below the ϒð2SÞ resonance.
The EVTGENevent generator[20]is used to simulate the
signal event decay chain, eþe−→ ϒð2SÞ → πþπ−ϒð1SÞ,
ϒð1SÞ → γA0, A0→ c¯c, for A0 masses between 4.0 and
9.0 GeV=c2in 0.5 GeV=c2steps and for A0masses of 9.2,
9.3, and 9.4 GeV=c2. The A0decay width is assumed to be
1 MeV. The hadronization of the c¯c system is simulated using the JETSET [21] program. The detector response is
simulated with the GEANT4 [22]suite of programs. Photon candidates are required to have an energy greater than 30 MeV and a Zernike moment A42[23]less than 0.1.
The A42 selection reduces contributions from hadronic
showers identified as photons. Events are required to contain at least one photon candidate. Each photon can-didate is taken in turn to represent the radiative photon in the ϒð1SÞ → γA0 decays. We do not select a best signal
candidate, neither for the radiative photon nor for the D meson and dipion candidates discussed below, but rather allow multiple candidates in an event.
Events must contain at least one D meson candidate, which is reconstructed in five channels: D0→ K−πþ,
Dþ → K−πþπþ, D0→ K−πþπþπ−, D0→ K0
Sπþπ−, and
D$ð2010Þþ→ πþD0 with D0→ K−πþπ0. The D0→
K−πþπ0 decays are reconstructed in the D$ð2010Þþ
pro-duction channel to reduce a large background that would otherwise be present. The inclusion of charge conjugate processes is implied. The π0 candidates are reconstructed
from two photon candidates by requiring the invariant mass of the reconstructed π0to lie between 100 and 160 MeV=c2.
The π0candidates do not make use of the radiative photon
candidate. The K0
S candidates are reconstructed from two
oppositely charged pion candidates. Each K0
Scandidate must
have a reconstructed mass within 25 MeV=c2of the nominal
K0
Smass[19]and satisfy d=σd> 3, where d is the distance
between the reconstructed eþe− collision point and the K0 S
vertex, with σdthe uncertainty of d.
The D0 and Dþ candidates are required to have
masses within 20MeV=c2 of their nominal masses [19],
corresponding to three to four standard deviations (σ) in their
mass resolution. When reconstructing D$ð2010Þþ
candi-dates, we constrain the D0→ K−πþπ0candidate mass to its
nominal value[19]. The D$ð2010Þþ candidate mass
distri-bution has longer tails. The D$ð2010Þþ candidates are
required to lie within 5 MeV=c2of its nominal mass[19],
corresponding to 10σ in the mass resolution.
Events are required to have at least one dipion candidate, constructed from two oppositely charged tracks. The invari-ant mass, mR, of the system recoiling against the dipion in the
ϒð2SÞ → πþπ−ϒð1SÞ transition is calculated by
m2
R ¼ M2ϒð2SÞþ m2ππ− 2Mϒð2SÞEππ; ð1Þ
where mππ is the measured dipion mass, Mϒð2SÞ is the
nominal ϒð2SÞ mass[19], and Eππ is the dipion energy in
the eþe− center-of-mass (CM) frame. The two pions in the
dipion system are required to arise from a common vertex. Signal candidates must satisfy 9.45 < mR< 9.47 GeV=c2.
Figure1presents the distribution of mRafter application of
these criteria. A clear peak is seen at the ϒð1SÞ mass. All charged tracks and calorimeter clusters other than those used to define the radiative photon, the D meson candidate, and the dipion candidate are referred to as the “rest of the event.”
The mass of the A0 candidate, m
X, is determined from
the mass of the system recoiling against the dipion and photon through
m2
X¼ ðPeþe−− Pπþπ−− PγÞ2; ð2Þ
where P denotes four-momentum measured in the CM frame. The four-momentum of the eþe− system is given
by Peþe− ¼ ðMϒð2SÞ; 0; 0; 0Þ.
Backgrounds are evaluated using simulated ϒð2SÞ and eþe− → q¯q events, where q is a u; d; s, or c quark. Events
with low-energy photons contribute a large background for mX greater than 7.50 GeV=c2. Therefore, the analysis is
) 2 (GeV/c R m 9.45 9.455 9.46 9.465 9.47 2 Candidates / MeV/c 0 500 1000 1500 3 10 × on-resonance data on-resonance luminosity off-resonance data normalized to
FIG. 1. The mRdistribution of events with a dipion, charm, and photon tag before application of selection criteria based on the BDT output (see text). The solid circles indicate the on-resonance data. The open squares indicate the off-resonance data normal-ized to the on-resonance luminosity.
divided into a low A0 mass region (4.00 to 8.00 GeV=c2)
and a high A0 mass region (7.50 to 9.25 GeV=c2). The
definitions of the regions, which overlap, are motivated by the need to have sufficient statistical precision for the background determination in each region.
We train ten boosted decision tree (BDT) classifiers[24]
to separate background from signal candidates (two mass regions × five D channels). The BDTs are trained using samples of simulated signal events, simulated generic ϒð2SÞ events, and the off-resonance data. The BDT inputs consist of 24 variables:
(1–2) Event variables:
(a) number of charged tracks in the event, (b) number of calorimeter clusters in the event. (3–12) Kinematic variables:
(a) mR,
(b) dipion likelihood (defined later), (c) D candidate mass,
(d) D candidate momentum, (e) photon π0score (defined later),
(f) energy of the most energetic charged track in the rest of the event, calculated using a charged pion mass hypothesis,
(g) energy of the most energetic calorimeter cluster in the rest of the event,
(h) invariant mass of the rest of the event,
(i) CM frame momentum of the rest of the event, (j) CM frame energy of the rest of the event. (13–15) Vertex variables:
(a) transverse coordinate of a vertex formed using all charged tracks,
(b) longitudinal coordinate of a vertex formed using all charged tracks,
(c) the χ2probability of a vertex fit using all charged
tracks.
(16–18) Event shape variables:
(a) the ratio of the second to zeroth Fox-Wolfram moment[25], calculated using all charged tracks and calorimeter clusters,
(b) sphericity [26]of the event, (c) magnitude of the thrust[27].
(19–24) Opening angles in the CM frame between the (a) dipion and photon candidate,
(b) dipion and D candidate, (c) dipion and thrust axis, (d) photon and D candidate, (e) photon and thrust axis, (f) D candidate and thrust axis.
The kinematic variables provide the most separation power for all ten BDTs. The separation power of the other variables depends on the mass region and channel. The vertex variables suppress background without a D meson in the event. The event shape variables suppress eþe− → q¯q
backgrounds.
The dipion likelihood[24]is defined using the opening angle between the two charged pions in the CM frame, the
transverse momentum of the pair, the invariant mass of the pair, the larger of the two momenta of the pair, and the χ2
probability of the pair’s vertex fit.
To reject photon candidates from π0→ γγ decays, a
likelihood[24] is defined using the invariant mass of the radiative photon candidate and a second photon (if present), and the second photon’s CM energy. The lower the like-lihood, the more π0 -like the photon pair. The photon π0
score is the minimum likelihood formed from the radiative photon and any other photon in the event excluding photon candidates used to reconstruct the π0candidate in the D0→
K−πþπ0 decay.
For each channel and mass range, each BDT output variable is required to exceed a value determined by maximizing the quantity S=ð0.5Nσþ
ffiffiffiffi B p
Þ [28], where S and B are the expected numbers of signal and background events, respectively, based on simulation, and Nσ ¼ 3 is the
number of standard deviations desired from the result. In the case of events with multiple signal candidates that satisfy the selection criteria, there may be multiple values of mX. Signal candidates that have the same dipion and
radiative photon candidate have the same value of mX,
irrespective of which D candidate is used. We reject a signal candidate if its value of mX has already been used.
In total, 9.8 × 103 and 7.4 × 106 candidates satisfy the
selection criteria in the low- and high-mass regions, respec-tively. The corresponding distributions of mX are shown in
Fig.2. The backgrounds in the low-mass region consist of ϒð1SÞ → γgg (35%), other ϒð1SÞ decays, denoted ϒð1SÞ → X (34%), ϒð2SÞ decays without a dipion transi-tion, denoted ϒð2SÞ → X (15%), and eþe− → q ¯q events
(16%). The corresponding background contributions in the high-mass region are 1%, 66%, 18%, and 15%. Background contributions from ϒð1SÞ → γgg decays reach a maximum near 5.5 GeV=c2and decrease above 7 GeV=c2.
We search for the A0 resonance as a peak in the m X
distribution. The reconstructed width of the A0is expected
to strongly depend on its mass due to better photon energy resolution at lower photon energies. Therefore, an extended maximum likelihood fit in a local mass range is performed as a function of test-mass values, denoted mA0. For these
fits, the parameters of the probability density function (PDF) used to model the shape of the signal distribution are fixed. The parameters of the background PDF, the number of signal events Nsig, and the number of background events
are determined in the fit.
The signal mX PDF is modeled with a Crystal Ball
function [29], which consists of a Gaussian and a power-law component. The values of the signal PDF at a given value of mA0 are determined through interpolation from fits of
simulated signal events at neighboring masses. The back-ground mXPDF is modeled with a second-order polynomial.
The fits are performed to the mX spectrum, for various
choices of mA0, in steps of 10 and 2 MeV=c2for the low- and
3 times smaller than the width of the signal mXPDF. We use a
local fitting range of %10σCBaround mA0, where σCBdenotes
the width of the Gaussian component of the Crystal Ball function. The σCB parameter varies between 120 and
8 MeV=c2 for values of m
A0 between 4.00 and
9.25 GeV=c2, as shown in Fig.3. We do not perform a fit
for 8.95 < mA0< 9.10 GeV=c2 because of a large
back-ground from ϒð2SÞ→γχbJð1PÞ, χbJð1PÞ→γϒð1SÞ decays.
The fitting procedure is validated using background-only pseudoexperiments. The mX PDF used to generate
pseu-doexperiments for the low-mass region is obtained from a fit of a fifth-order polynomial to the low-mass region data. The mX PDF used for the high-mass region is obtained
from a fit of the sum of four exponential functions plus six Crystal Ball functions to the high-mass region data, with shape parameters fixed according to expectations from simulation and with the relative normalizations determined in the fit. The Crystal Ball functions describe
the ϒð2SÞ → γχbJð1PÞ and χbJð1PÞ → γϒð1SÞ transitions
while the exponential terms describe the nonresonant background. Four exponential terms are used because the nonresonant background increases rapidly for higher mX. The background fits are overlaid in Fig.2. The fitting
procedure returns a null signal for most mA0 values but is
found to require a correction to Nsigfor values of mA0near
4.00 or 9.25 GeV=c2. The corrections are determined from
the average number of signal events found in the fits to the background-only pseudoexperiments. The corrections are applied as a function of mA0 and reach a maximum of 15
and 50 candidates in the low- and high-mass regions, respectively. The uncertainty of the correction is assumed to be half its value.
The reconstruction efficiency takes into account the hadronization of the c¯c system into D mesons, the branching fraction of D mesons to the five decay channels, detector acceptance, and the BDT selection. The efficien-cies range from 4.0% to 2.6% for simulated A0 masses
between 4.00 and 9.25 GeV=c2.
Potential bias introduced by the fitting procedure is evaluated using pseudoexperiments with different values of the product branching fraction Bðϒð1SÞ → γA0
Þ× BðA0→ c¯c). For various choices of m
A0, the extracted
product branching fraction is found to be ð4 % 3Þ% higher than the value used to generate the events. This result is used to define a correction and its uncertainty.
TableIsummarizes all correction factors and associated systematic uncertainties. The fit correction systematic uncertainty is added in quadrature with the statistical uncertainty of Nsig. All other correction factors are added
in quadrature and applied to the reconstruction efficiency. A correction of 1.00 means we do not apply any correction but propagate the multiplicative uncertainty.
The systematic uncertainties associated with the reconstruction efficiencies are dominated by the differences between data and simulation, including the BDT output modeling, c¯c hadronization, D-candidate mass resolu-tion, dipion recoil mass and likelihood modeling, and photon reconstruction. Other systematic uncertainties are
) 2 mass (GeV/c 0 Simulated A 4 5 6 7 8 9 ) 2 (GeV/c CB σ 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
FIG. 3. The σCBparameter for A0decays of various simulated masses. ) 2 (GeV/c X m 3 4 5 6 7 8 2 Candidates / 100 MeV/c 0 200 400 600 800 1000 (a) data fit to data gg γ → (1S) Υ X → (1S) Υ X → (2S) Υ q q → -e + e ) 2 (GeV/c X m 7.5 8 8.5 9 2 Candidates / 10 MeV/c 10 2 10 3 10 4 10 (b)
FIG. 2 (color online). The mXdistributions of signal candidates in the low- (a) and high- (b) mass regions after applying all selection criteria. The points indicate the data. The solid curve shows the result of a fit to the data under a background-only hypothesis. The colored histograms show the cumulative back-ground contributions from eþe−→ q¯q (magenta dense-dot filled), ϒð2SÞ → X (green sparse-dot filled), ϒð1SÞ → X (blue dotted), and ϒð1SÞ → γgg (red dashed) events.
associated with the fit bias (discussed above), the dipion branching fraction [19], the finite size of the simulated signal sample, and the ϒð2SÞ counting [18].
The BDT output distributions in off-resonance data and eþe−→ q¯q simulation, shown in Fig. 4, have consistent
shapes but are slightly shifted from one another. The associated systematic uncertainty is estimated by shifting the simulated distributions so that the mean values agree with the data, and then recalculating the efficiencies. The reconstruction efficiencies decrease by 7% and 2% in the low- and high-mass regions, respectively.
The uncertainty associated with c¯c hadronization is evaluated by comparing D meson production in off-resonance data and eþe− → c¯c simulation normalized to
the same luminosity. The difference in the yield varies from 1% to 9% for the five D decay channels. We conservatively assign a global multiplicative uncertainty of 9% that includes effects due to the hadronization modeling, particle identification, tracking, π0 reconstruction, and luminosity
determination of the off-resonance data.
The uncertainty due to the discrepancy between the reconstructed D mass resolution in data and simulation is estimated by Gaussian smearing of the D mass input in simulation to match the data and measuring the difference in the reconstruction efficiency.
Further corrections to account for data and simulation differences in reconstruction efficiencies are estimated with similar methods. Corrections are applied to account for the dipion recoil mass reconstruction, the dipion likelihood modeling, and the photon reconstruction[30].
The highest observed local significance in the low-mass region is 2.3 standard deviations, including statistical uncertainties only, at 4.145 GeV=c2. The corresponding
result for the high-mass region is 2.0 standard deviations at 8.411 GeV=c2. The fits are shown in Fig. 5. Such
) 2 (GeV/c X m 3 3.5 4 4.5 5 2 Candidates / 0.05 GeV/c 0 20 40 60 80 100 2 Candidates / 0.05 GeV/c (a) 2 5 5 8 4 6 4 4 4 2 4 4 8 3 6 3 4 3 2 3 3 Data -2 0 2 2 5 5 8 4 6 4 4 4 2 4 4 8 3 6 3 4 3 2 3 3 Data-Fit ) 2 (GeV/c X m 8.2 8.3 8.4 8.5 8.6 2 Candidates / 0.01 GeV/c 0 200 400 600 800 1000 1200 2 Candidates / 0.01 GeV/c (b) Data 0 2 Data-Fit -2
FIG. 5. The fits with the highest local significance in the low-(a) and high- (b) mass regions. The solid line is the fit that includes a signal. The dotted line is the background component of the solid line.
TABLE I. Summary of corrections and their associated sys-tematic uncertainties. All corrections are multiplicative except for the fit correction.
Source Low region High region
Fit correction (candidates) Up to 15 % 8 Up to 50 % 25 BDT output modeling 0.93% 0.04 0.98% 0.01
Source Both regions
c¯c hadronization 1.00% 0.09
Fit bias 1.04% 0.03
Dipion branching fraction 1.00% 0.02
Photon reconstruction 0.967% 0.017
D mass resolution 0.98% 0.01
Finite simulation statistics 1.00% 0.01
ϒð2SÞ counting 1.00% 0.01
Dipion likelihood 1.02% 0.01
Dipion recoil mass 0.991% 0.005
100 200 (a) Kπ 10000 20000 (b) Kπ 200 400 (c) Kππ 20000 40000 60000 (d) Kππ 500 1000 (e) Kπππ 100000 200000 (f) Kπππ 100 200 (g) ππ S K 10000 20000 (h) ππ S K -1 -0.5 0 0.5 0 50 100 (i) 0 ππ K -1 -0.5 0 0.5 0 10000 20000 (j) Kππ0 BDT output Candidates / 0.1
FIG. 4. The BDT distributions in off-resonance data (points) and simulated eþe−→ q¯q events (histograms) for the five D meson decay modes. The results on the left (a, c, e, g, i) and right (b, d, f, h, j) correspond to the low- and high-mass regions, respectively.
fluctuations occur in 54% and 80% of pseudoexperiments, respectively. Hence, our data are consistent with the background-only hypothesis.
Upper limits on the product branching fraction Bðϒð1SÞ → γA0Þ × BðA0→ c¯cÞ at 90% confidence level
(C.L.) are determined assuming a uniform prior, with the constraint that the product branching fraction be greater than zero. The distribution of the likelihood function for Nsig is assumed to be Gaussian with a width equal to the
total uncertainty in Nsig. The upper limits obtained from the
low-mass region are combined with those from the high-mass region to define a continuous spectrum for the upper limits. The results are shown in Fig. 6.
In summary, we search for a resonance in radiative decays of the ϒð1SÞ with a charm tag. We do not observe a significant signal and set upper limits on the product branching fraction Bðϒð1SÞ → γA0
Þ × BðA0→ c¯cÞ
rang-ing from 7.4 × 10−5 to 2.4 × 10−3 for A0 masses from
4.00 to 9.25 GeV=c2, excluding masses from 8.95 to
9.10 GeV=c2 because of background from ϒð2SÞ →
γχbJð1PÞ, χbJð1PÞ → γϒð1SÞ decays. These results will
further constrain the NMSSM parameter space.
We are grateful for the extraordinary contributions of our PEP-II colleagues in achieving the excellent luminosity and machine conditions that have made this work possible. The success of this project also relies critically on the expertise and dedication of the computing organizations that support BABAR. The collaborating institutions wish to thank SLAC for its support and the kind hospitality extended to them. This work is supported by the U.S. Department of Energy and National Science Foundation, the Natural Sciences and Engineering Research Council (Canada), the Commissariat à l’Energie Atomique and Institut National de Physique Nucléaire et de Physique des Particules (France), the Bundesministerium für Bildung und Forschung and Deutsche Forschungsgemeinschaft (Germany), the Istituto Nazionale di Fisica Nucleare (Italy), the Foundation for Fundamental Research on Matter (The Netherlands), the Research Council of Norway, the Ministry of Education and Science of the Russian Federation, Ministerio de Economía y Competitividad (Spain), the Science and Technology Facilities Council (United Kingdom), and the Binational Science Foundation (U.S.-Israel). Individuals have received support from the Marie-Curie IEF program (European Union) and the A. P. Sloan Foundation (USA).
[1] R. Dermisek, J. F. Gunion, and B. McElrath,Phys. Rev. D 76, 051105(R) (2007).
[2] M. Maniatis, Int. J. Mod. Phys. A 25, 3505 (2010). [3] G. Aad et al. (ATLAS Collaboration),Phys. Lett. B 716, 1
(2012).
[4] S. Chatrchyan et al. (CMS Collaboration), Phys. Lett. B 716, 30 (2012).
[5] N. D. Christensen, T. Han, Z. Liu, and S. Su,J. High Energy Phys. 08 (2013) 019.
[6] F. Wilczek,Phys. Rev. Lett. 39, 1304 (1977).
[7] R. Dermisek and J. F. Gunion, Phys. Rev. D 81, 075003 (2010).
[8] B. Aubert et al. (BABAR Collaboration), Phys. Rev. Lett. 103, 081803 (2009).
[9] J. P. Lees et al. (BABAR Collaboration),Phys. Rev. D 87, 031102(R) (2013).
[10] J. P. Lees et al. (BABAR Collaboration),Phys. Rev. D 88, 031701(R) (2013).
[11] B. Aubert et al. (BABAR Collaboration),Phys. Rev. Lett. 103, 181801 (2009).
[12] J. P. Lees et al. (BABAR Collaboration),Phys. Rev. D 88, 071102(R) (2013).
[13] S. Chatrchyan et al. (CMS Collaboration),Phys. Rev. Lett. 109, 121801 (2012).
[14] S. Chatrchyan et al. (CMS Collaboration), Phys. Lett. B 726, 564 (2013).
[15] W. Love et al. (CLEO Collaboration),Phys. Rev. Lett. 101, 151802 (2008).
[16] B. Aubert et al. (BABAR Collaboration), Nucl. Instrum. Methods Phys. Res., Sect. A 479, 1 (2002).
[17] B. Aubert et al. (BABAR Collaboration), Nucl. Instrum. Methods Phys. Res., Sect. A 729, 615 (2013).
) 2 (GeV/c 0 A m 4 5 6 7 8 9 90% C.L. upper limits 10-4 -3 10
Expected average Observed limits
Expected (68%) (statistical uncertainties only)Observed limits
Expected (95%)
FIG. 6 (color online). The 90% C.L. upper limits on the product branching fraction Bðϒð1SÞ → γA0Þ × BðA0→ c¯cÞ using all uncertainties (thick line) and using statistical uncertainties only (thin dashed line). The inner and outer bands contain 68% and 95% of our expected upper limits. The bands are calculated using all uncertainties. The thin solid line in the center of the inner band is the expected upper limit.
[18] J. P. Lees et al. (BABAR Collaboration), Nucl. Instrum. Methods Phys. Res., Sect. A 726, 203 (2013).
[19] K. A. Olive et al. (Particle Data Group),Chin. Phys. C 38, 090001 (2014).
[20] D. J. Lange, Nucl. Instrum. Methods Phys. Res., Sect. A 462, 152 (2001).
[21] T. Sjöstrand,Comput. Phys. Commun. 82, 74 (1994). [22] S. Agostinelli et al. (GEANT4 Collaboration), Nucl.
Ins-trum. Methods Phys. Res., Sect. A 506, 250 (2003). [23] R. Sinkus and T.Voss,Nucl. Instrum. Methods Phys. Res.,
Sect. A 391, 360 (1997).
[24] A. Höcker et al., Proc. Sci., ACAT2007 (2007) 040. [25] G. C. Fox and S. Wolfram, Nucl. Phys. B149, 413
(1979).
[26] J. Bjorken and S. Brodsky,Phys. Rev. D 1, 1416 (1970). [27] S. Brandt, Ch. Peyrou, R. Sosnowski, and A. Wroblewski,
Phys. Lett. 12, 57 (1964).
[28] G. Punzi, Proceedings of PHYSTAT (2003), p. 79. [29] M. J. Oreglia, Ph.D. thesis, Report No. SLAC-R-236, 1980,
Appendix D.
[30] P. del Amo Sanchez et al. (BABAR Collaboration), Phys. Rev. Lett. 107, 021804 (2011).
